All entries

Characterizing the Robustness of Black-Box LLM Planners Under Perturbed Observations with Adaptive Stress Testing

arXiv:2505.05665v4 Announce Type: replace-cross Abstract: Large language models (LLMs) have recently demonstrated success in decision-making tasks including planning, control, and prediction, but thei

safetyarxiv-cs-cl

ChatGPT used the term “human slop”

DGX agent

The term 'human slop' refers to low-effort content created by humans, sometimes harder to detect than AI-generated content because it fails to expose itself immediately. The concept emerged as a count

industryr-chatgpt

13 May 2026

Checkup2Action: A Multimodal Clinical Check-up Report Dataset for Patient-Oriented Action Card Generation

model-releasesarxiv-cs-cl

arXiv:2605.11533v1 Announce Type: new Abstract: Clinical check-up reports are multimodal documents that combine page layouts, tables, numerical biomarkers, abnormality flags, imaging findings, and dom

CheXTemporal: A Dataset for Temporally-Grounded Reasoning in Chest Radiography

arXiv:2605.11304v1 Announce Type: new Abstract: Chest radiograph interpretation requires temporal reasoning over prior and current studies, yet most vision-language models are trained on static image-

safetyarxiv-cs-cv

China's CSI AI Index, which tracks AI supply chain companies, is up 28%+ YTD, while HK's Hang Seng Tech Index, which includes Alibaba and Tencent, is down 8% (William Sandlund/Financial Times)

William Sandlund / Financial Times: China's CSI AI Index, which tracks AI supply chain companies, is up 28%+ YTD, while HK's Hang Seng Tech Index, which includes Alibaba and Tencent, is down 8% — Tenc

industrytechmeme

China's move to block Meta's Manus acquisition challenges the 'Singapore washing' model, as major Chinese tech companies build significant Singapore presences (Owen Walker/Financial Times)

Owen Walker / Financial Times: China's move to block Meta's Manus acquisition challenges the “Singapore washing” model, as major Chinese tech companies build significant Singapore presences — Beijing'

industrytechmeme

Choosing features for classifying multiword expressions

arXiv:2605.11779v1 Announce Type: new Abstract: Multiword expressions (MWEs) are a heterogeneous set with a glaring need for classifications. Designing a satisfactory classification involves choosing

researcharxiv-cs-cl

Chronicles-OCR: A Cross-Temporal Perception Benchmark for the Evolutionary Trajectory of Chinese Characters

model-releasesarxiv-cs-cv

arXiv:2605.11960v1 Announce Type: new Abstract: Vision Large Language Models (VLLMs) have achieved remarkable success in modern text-rich visual understanding. However, their perceptual robustness in

Hardware

ChunkFlow: Communication-Aware Chunked Prefetching for Layerwise Offloading in Distributed Diffusion Transformer Inference

arXiv:2605.11335v1 Announce Type: cross Abstract: Layerwise offloading reduces the GPU memory footprint of large diffusion transformer (DiT) inference by prefetching upcoming layers from host memory,

hardwarearxiv-cs-lg

Cisco reports Q3 revenue up 12% YoY to 15.84B, vs. 15.56B est., forecasts Q4 revenue above est., and is cutting almost 4,000 jobs; CSCO jumps 17%+ after hours (Jordan Novet/CNBC)

Jordan Novet / CNBC: Cisco reports Q3 revenue up 12% YoY to 15.84B, vs. 15.56B est., forecasts Q4 revenue above est., and is cutting almost 4,000 jobs; CSCO jumps 17%+ after hours — Cisco shares soare

industrytechmeme

clarifying how much “task horizon” falls off as a function of the increasing accuracy criterion directly in the graph (not across tabs) woul…

model-releasesgary-marcus--x

clarifying how much “task horizon” falls off as a function of the increasing accuracy criterion directly in the graph (not across tabs) would definitely improve this graph, @METR_Evals. as a variant o

Clarity: The Flexibility-Interpretability Trade-Off in Sparsity-aware Concept Bottleneck Models

arXiv:2601.21944v2 Announce Type: replace Abstract: The widespread adoption of deep learning models in computer vision has intensified concerns about interpretability. Despite strong performance, thes

safetyarxiv-cs-lg

Claude Code and @suno have more in common than you might think: 'It's fun to build things, and it's fun to use what you build.' AI lets peop…

model-releasessonya-huang--x

Claude Code and @suno have more in common than you might think: 'It's fun to build things, and it's fun to use what you build.' AI lets people be creative in almost any domain, from coding to making m

Claude Code weekly limits are increasing 50%, now through July 13. Live now for all Pro, Max, Team, and seat-based Enterprise users.

model-releasesboris-cherny--x

Claude Code weekly usage limits have increased by 50% for all Pro, Max, Team, and seat-based Enterprise users, effective immediately through July 13. This temporary expansion applies across all user t

Clinical operations intelligence belongs on the Lakehouse

Clinical operations intelligence in healthcare can be effectively managed and analyzed using a lakehouse architecture, which combines the benefits of data lakes and data warehouses. This approach enab

industrydatabricks

ClinicalBench: Stress-Testing Assertion-Aware Retrieval for Cross-Admission Clinical QA on MIMIC-IV

model-releasesarxiv-cs-cl

arXiv:2605.11143v1 Announce Type: new Abstract: Reasoning benchmarks measure clinical performance on clean inputs. We evaluate the step before reasoning: retrieval over real EHR notes, where negation,

Closing the Motion Execution Gap: From Semantic Motion Task Constraints to Kinematic Control

arXiv:2605.12053v1 Announce Type: new Abstract: This paper addresses the Motion Execution Gap, the disconnect between high-level symbolic task descriptions using semantic constraints and executable ro

researcharxiv-cs-ro

Cluster-Aware Neural Collapse Prompt Tuning for Long-Tailed Generalization of Vision-Language Models

arXiv:2605.11939v1 Announce Type: new Abstract: Prompt learning has emerged as an efficient alternative to fine-tuning pre-trained vision-language models (VLMs). Despite its promise, current methods s

safetyarxiv-cs-cv

Hardware

CME Group and Silicon Data to launch AI compute futures market

Silicon Data, the startup that provides market intelligence for artificial intelligence compute infrastructure, will provide the price indexes for a new futures market that will allow investors to hed

hardwaresiliconangle

Co-sign.

Co-sign. We asked the CEO of HuggingFace @ClementDelangue what the risks of releasing powerful open source models are. He says restricting AI creates more risk than openness. 'Six, seven years ago, at

industryclem-delangue--x

codex is the best AI coding product and we want to make it easy to try. for the next 30 days, we are giving companies that want to try switc…

OpenAI announced a 30-day promotional period offering companies easy access to try Codex, positioning it as a leading AI coding product. The initiative aims to lower barriers for businesses interested

industrysam-altman--x

Coevolutionary Continuous Discrete Diffusion: Make Your Diffusion Language Model a Latent Reasoner

arXiv:2510.03206v2 Announce Type: replace-cross Abstract: Diffusion language models, especially masked discrete diffusion models, have achieved great success recently. While there are some theoretical

applicationsarxiv-cs-cl

Cofense adds AI-driven campaign detection to its phishing defense platform

Phishing defense company Cofense Inc. today unveiled new artificial intelligence-driven detection and automation features for its Phishing Defense Platform aimed at catching coordinated polymorphic ph

industrysiliconangle

Combining On-Policy Optimization and Distillation for Long-Context Reasoning in Large Language Models

arXiv:2605.12227v1 Announce Type: new Abstract: Adapting large language models (LLMs) to long-context tasks requires post-training methods that remain accurate and coherent over thousands of tokens. E

safetyarxiv-cs-cl

Agents

ComfyUI Skill in Hermes Agent Deep Dive https://x.com/i/broadcasts/1NGaraWmzBdJj

This broadcast likely provides an in-depth technical discussion of how to implement and utilize ComfyUI skills within the Hermes Agent framework, covering integration methods, practical applications,

agentscomfyui--x

Agents

Coming up at Interrupt: 🎙️ How @clay scales their GTM Engineering agents with Head of AI @JeffBarg. 🎙️ @ummadisetti and @kordelfrance on h…

Coming up at Interrupt: 🎙️ How @clay scales their GTM Engineering agents with Head of AI @JeffBarg. 🎙️ @ummadisetti and @kordelfrance on how @Toyota Motor North America equipped 56,000 employees with

agentsharrison-chase--x

Compositional Neural Operators for Multi-Dimensional Fluid Dynamics

arXiv:2605.11691v1 Announce Type: new Abstract: Partial differential equations (PDEs) govern diverse physical phenomena, yet high-fidelity numerical solutions are computationally expensive and Machine

researcharxiv-cs-lg

Computational Design of a Low-Visibility UAV Using a Human-Aligned Perceptual Metric

arXiv:2605.11296v1 Announce Type: new Abstract: We introduce Phantom Twist, a type of single-propeller UAV designed to achieve low visibility through high-speed spinning and the exploitation of motion

researcharxiv-cs-ro

Tools

Computer is secure by default. Every task runs in its own hardware-isolated sandbox with VPC-level storage and compute separation. Agents ar…

Computer is secure by default. Every task runs in its own hardware-isolated sandbox with VPC-level storage and compute separation. Agents are authenticated with short-lived proxy tokens instead of raw

toolsperplexity--x

Concepts in Motion: Temporal Concept Bottleneck Model for Interpretable Video Classification

arXiv:2509.20899v3 Announce Type: replace Abstract: Concept Bottleneck Models (CBMs) enable interpretable image classification by structuring predictions around human-understandable concepts, but exte

researcharxiv-cs-cv

Concordance Comparison as a Means of Assembling Local Grammars

arXiv:2605.11862v1 Announce Type: new Abstract: Named Entity Recognition for person names is an important but non-trivial task in information extraction. This article uses a tool that compares the con

applicationsarxiv-cs-cl

Config, which is building a data layer for robotics foundation models, raised a 27M seed at a 200M+ valuation led by Samsung Venture Investment (Kate Park/TechCrunch)

Kate Park / TechCrunch: Config, which is building a data layer for robotics foundation models, raised a 27M seed at a 200M+ valuation led by Samsung Venture Investment — Asia's push into physical AI i

applicationstechmeme

Constrained Stochastic Spectral Preconditioning Converges for Nonconvex Objectives

arXiv:2605.11850v1 Announce Type: cross Abstract: In this work, we develop proximal preconditioned gradient methods with a focus on spectral gradient methods providing a proximal extension to the Muon

researcharxiv-cs-lg

Context Convergence Improves Answering Inferential Questions

arXiv:2605.12370v1 Announce Type: new Abstract: While Large Language Models (LLMs) are widely used in open-domain Question Answering (QA), their ability to handle inferential questions-where answers m

researcharxiv-cs-cl

Context Steering: A New Paradigm for Compression-based Embeddings by Synthesizing Relevant Information Features

arXiv:2508.14780v2 Announce Type: replace Abstract: Compression-based dissimilarities (CD) offer a flexible and domain-agnostic means of measuring similarity by identifying implicit information throug

applicationsarxiv-cs-lg

Continuous Discovery of Vulnerabilities in LLM Serving Systems with Fuzzing

arXiv:2605.11202v1 Announce Type: cross Abstract: LLM inference and serving systems have become security-critical infrastructure; however, many of their most concerning failures arise from the serving

safetyarxiv-cs-lg

Contrastive Learning under Noisy Temporal Self-Supervision for Colonoscopy Videos

arXiv:2605.12320v1 Announce Type: new Abstract: Learning robust representations of polyp tracklets is key to enabling multiple AI-assisted colonoscopy applications, from polyp characterization to auto

researcharxiv-cs-cv

Agents

Control Charts for Multi-agent Systems

arXiv:2605.11135v1 Announce Type: cross Abstract: Generative agents have proven to be powerful assistants in a wide variety of contexts. Given this success, users are now deploying agents with minimal

agentsarxiv-cs-lg

Control of Fully Actuated Aerial Vehicles: A Comparison of Model-based and Sensor-based Dynamic Inversion

model-releasesarxiv-cs-ro

arXiv:2605.12071v1 Announce Type: new Abstract: Fully actuated multirotor platforms decouple translational force generation from vehicle attitude, enabling independent control of position and orientat

Controllable User Simulation

arXiv:2605.11519v1 Announce Type: cross Abstract: Using offline datasets to evaluate conversational agents often fails to cover rare scenarios or to support testing new policies. This has motivated th

safetyarxiv-cs-cl

Tutorials

cool launch

tutorialsyohei-nakajima--x

cool launch Excited to co-found Recursive (@recursive_si) with an exceptional team in London and SF to create AI that experiments on how to safely improve itself, turning compute into knowledge that a

Cooperative Robotics Reinforced by Collective Perception for Traffic Moderation

arXiv:2605.11972v1 Announce Type: new Abstract: Collisions at non-line-of-sight (NLOS) intersections remain a major safety concern because drivers have limited visibility of approaching traffic. V2X b

safetyarxiv-cs-ro

Coordinated Diffusion: Generating Multi-Agent Behavior Without Multi-Agent Demonstrations

arXiv:2605.11485v1 Announce Type: new Abstract: Imitation learning powered by generative models has proven effective for modeling complex single-agent behaviors. However, teaching multi-agent systems,

safetyarxiv-cs-ro

CORE: Cyclic Orthotope Relation Embedding for Knowledge Graph Completion

model-releasesarxiv-cs-lg

arXiv:2605.11159v1 Announce Type: new Abstract: Knowledge graph completion (KGC) aims to automatically infer missing facts in multi-relational data by mapping entities and relations into continuous re

Correcting Selection Bias in Sparse User Feedback for Large Language Model Quality Estimation: A Multi-Agent Hierarchical Bayesian Approach

model-releasesarxiv-cs-cl

arXiv:2605.12177v1 Announce Type: new Abstract: [Abridged] Production LLM deployments receive feedback from a non-random fraction of users: thumbs sit mostly in the tails of the satisfaction distribut

Local Ai

COSMOS: Model-Agnostic Personalized Federated Learning with Clustered Server Models and Pseudo-Label-Only Communication

arXiv:2605.11165v1 Announce Type: new Abstract: Federated learning (FL) in heterogeneous environments remains challenging because client models often differ in both architecture and data distribution.

local-aiarxiv-cs-lg

Could this be the moment that drug manufacturing takes off in orbit?

Varda Space Industries and United Therapeutics announced a collaboration to conduct pharmaceutical processing of small molecule medicines in orbit, combining Varda's orbital manufacturing platform wit

applicationsars-technica

counterpoint: i have no interest in a company that says we will roll everything out to anybody regardless of how dangerous it is.

counterpoint: i have no interest in a company that says we will roll everything out to anybody regardless of how dangerous it is. boris cherny recently says that, 'anthropic has no plans to roll out m

safetygary-marcus--x